Training and transfer of class inclusion in young children

Abstract

The main purposes of this study were: (1) to investigate the effectiveness of training methods and materials in the acquisition of the class inclusion concept, among five- and six-year old kindergarten and grade one children; (2) to examine the nature of young children's understanding, retention, and transfer of this logical ability; and (3) to examine the results in light of a cross-cultural framework which may have potential in interpreting responses from children in a different culture, socioeconomic level or simply a different setting. Piaget's theory indicates that the logic of classes and hierarchies is incomprehensible to preoperational children ~nti1 they have the abil ity to use the logical quantifiers "all" and "some". Subjects in this study were required to have reached the appropriate stage of "all" and "some" in a pretest designed for that purpose. A second pretest using Piaget's classic class inclusion experiment with wooden beads was administered to assess subjects' level of understanding class inclusion. Subjects were required to show lack of class inclusion understanding via the second pretest, in order to qualify for training. These two pretests were used to select sixty subjects who understood "all" and "some" but did not understand class inclusion. These subjects were then randomly assigned to six

treatment groups of ten subjects each, one of which was the control group. Five training conditions were designed, which combined methods "' and materials thus: self-discovery methods and concrete materials; se1fiv discovery methods and pictorial materials; tutorial methods and concrete materials; tutorial methods and pictorial materials; and verbal methods and materials. The control group received no training. Treatment groups received 10 to 30 minutes training designed to help them comprehend that when subclasses A and AI additively compose the superordinate class B, then B>A and B - AI = A. Four posttests from Immediate Transfer to One Month Transfer were administered at approximate ten-day intervals after training. One-Way Analysis of Variance tests were carried out and significant F ratios obtained between treatment groups and the control group. Multiple comparisons of mean scores were carried out using the studentized range statistic - Newman-Keuls method, and statistically significant Q values obtained between each of the treatment groups and the control group. When each posttest was considered separately as a dependent variable, there were no statistically significant differences among the means of the treatment groups (materials), except that all treatment groups were consistently superior to the control group. Similarly, there were no statistically significant differences among the treatment groups (methods). But, when all eight posttests were considered together as dependent variables, there were consistent differences among the treatment groups (materials) as the means for the groups were ranked by the eight posttests, even when the control~roup was omitted. Specifically, concrete materials . were consistently superior to pictorial materials. Similarly, tutorial methods were consist~ntly superior to self~discovery methods.

The main findings of this study showed that the training program was effective in inducing class inclusion among five- and six-year old middle-class, urban, Alberta children, when the appropriate training techniques were employed and when the appropriate stage of understanding lIallll and IIsome IIhad been reached. A significant finding of this study was the fact that the same conclusions were reached when scores with and without justification were used in the analyses. This finding has important implications for research in cross-cultural settings in which inferences based on the verbal ization of subjects' responses have led to biased results and erroneous conclusions on lack of competence for cognitive reasoning. These findings imply that the acquisition of class inclusion can be accelerated if the appropriate conditions are present. The use of a variety of materials and diverse teaching techniques in the classroom setting are suggested. For the planned replication study in Kenya (Phase II) less emphasis will be laid on the verbal justification criterion, but the use of concrete and familiar materials will be emphasized.